Capital Structure Katharina Lewellen Finance Theory II February 18 and 19, 2003 The Key Questions of Corporate Finance Valuation: How do we distinguish between good investment projects and bad ones? Financing: How should we finance the investment projects we choose to undertake? 2 (Real) Investment Policy “Which projects should the firm undertake?” ¾ ¾ ¾ ¾ Open a new plant? Increase R&D? Scale operations up or down? Acquire another company? We know that real investments can create value ¾ Discounted Cash Flow (DCF) analysis ¾ Positive NPV projects add value ¾ We revisit this in the course’s “Valuation” module (Part II) 3 Financing Policy Real investment policies imply funding needs ¾ We have tools to forecast the funding needs to follow a given real investment policy (from Wilson Lumber) But what is the best source of funds? ¾ Internal funds (i.e., Cash)? ¾ Debt (i.e., borrowing)? ¾ Equity (i.e., issuing stock)? Moreover, different kinds of ... ¾ Internal funds (e.g., cash reserves vs. cutting dividends) ¾ Debt (e.g., Banks vs. Bonds) ¾ Equity (e.g., VC vs. IPO) 4 Choosing an Optimal Capital Structure Is there an “optimal” capital structure, i.e., an optimal mix between debt and equity? More generally, can you add value on the RHS of the balance sheet, i.e., by following a good financial policy? If yes, does the optimal financial policy depend on the firm’s operations (Real Investment policy), and how? We study this in the course’s “Financing” module (Part I). 5 Capital Structures: US Corporations 1975-2001 Book leverage Market leverage debt / (debt + equity) (%) 50 40 30 20 10 0 1975 1980 1985 1990 1995 2000 6 Capital structure, International 1991 Book leverage Market leverage Debt / (Debt+Equity) (%) 60 50 40 30 20 10 0 US Japan UK Canada France Germany 7 Sources of Funds: US Corporations 1980-2000 int 120 debt equity % of total financing 100 80 60 40 20 0 -20 1980 1985 1990 1995 2000 -40 8 Sources of Funds: International 1990-94 Internal Internal 90 120 Debt Debt Equity Equity 80 100 % of total financing 70 80 60 60 50 40 30 20 10 40 20 0 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 -20 0 -40 US Japan UK Canada France 9 Examples: Capital structure, 1997 Industry High leverage Building construction Hotels and lodging Air transport Primary metals Paper Low leverage Drugs and chemicals Electronics Management services Computers Health services Debt / (Debt + Equity) (%) 60.2 55.4 38.8 29.1 28.2 4.8 9.1 12.3 9.6 15.2 10 Plan of Attack 1. Modigliani-Miller Theorem: → Capital Structure is irrelevant 2. What’s missing from the M-M view? → Taxes → Costs of financial distress 3. “Textbook” view of optimal capital structure: → The choice between debt and equity 4. Apply/confront this framework to several business cases → Evaluate when its usefulness and its limitations 11 M-M’s “Irrelevance” Theorem Assume Market efficiency and no asymmetric information No taxes No transaction or bankruptcy costs Hold constant the firm’s investment policies Then The value of the firm is independent of its capital structure ¾ Financing decisions do not matter! 12 MM Theorem: Proof 1 (pie theory)* Equity Equity Debt = Debt * Credit to Yogi Berra 13 MM Theorem: Proof 2 (market efficiency) Your firm decides to raise $100 million. Debt financing ¾ You sell bonds worth $100 million and receive $100 million in cash. Equity financing ¾ You sell stock worth $100 million and receive $100 million in cash. 14 MM Theorem: Proof 2 (market efficiency) All purely financial transactions are zero NPV investments, i.e., no arbitrage opportunity. Thus, they neither increase nor decrease firm value. 15 MM Theorem: Example Current Debt $200M Assets $1 billion Equity $800M Issue new debt Assets $1.1 billion Old Debt $200M New Debt $100M Equity $800M Issue new equity Assets $1.1 billion Debt $200M Old Eq $800M New Eq $100M 16 MM Theorem: Proof 3 Consider two firms with identical assets (in $M): Asset value next year: Firm A Firm B In state 1: In state 2: 160 40 160 40 Firm A is all equity financed: ¾ Firm A’s value is V(A) = E(A) Firm B is financed with a mix of debt and equity: ¾ Debt with one year maturity and face value $60M ¾ Market values of debt D(B) and equity E(B) ¾ Firm B’s value is (by definition) V(B) = D(B) + E(B) MM says: V(A) = V(B) 17 MM Theorem: Proof 3 Firm A’s equity gets all cash flows Firm B’s cash flows are split between its debt and equity with debt being senior to equity. Claim’s value next year In state 1: In state 2: Firm A (Equity) 160 40 Firm B Debt 60 40 Equity 100 0 In all (i.e., both) states of the world, the following are equal: ¾ The payoff to Firm A’s equity ¾ The sum of payoffs to Firm B’s debt and equity By value additivity, E(A) = D(B) + E(B) 18 M-M Intuition 1 If Firm A were to adopt Firm B’s capital structure, its total value would not be affected (and vice versa). This is because ultimately, its value is that of the cash flows generated by its operating assets (e.g., plant and inventories). The firm’s financial policy divides up this cashflow “pie” among different claimants (e.g., debtholders and equityholders). But the size (i.e., value) of the pie is independent of how the pie is divided up. 19 Example, cont. In case you forgot where value additivity comes from… Assume for instance that market values are: → D(B) = $50M → E(B) = $50M MM says: V(A) = D(B)+E(B) = $100M Suppose instead that E(A) = $105M. Can you spot an arbitrage opportunity? 20 Example, cont. Arbitrage strategy: ¾ Buy 1/1M of Firm B’s equity for $50 ¾ Buy 1/1M of Firm B’s debt for $50 ¾ Sell 1/1M of Firm A’s equity for $105 Today Firm B’s equity Firm B’s debt Subtotal Firm A’s equity Total -$50 -$50 -$100 +$105 +$5 Next year State 1 +$100 +$60 +$160 -$160 $0 Next year State 2 $0 +$40 +$40 -$40 $0 D Note: Combining Firm B’s debt and equity amounts to “undoing Firm B’s leverage” (see shaded cells). 21 M-M: Intuition 2 Investors will not pay a premium for firms that undertake financial transactions that they can undertake themselves (at the same cost). For instance, they will not pay a premium for Firm A over Firm B for having less debt. Indeed, by combining Firm B’s debt and equity in appropriate proportions, any investor can in effect “unlever” Firm B and reproduce the cashflow of Firm A. 22 The Curse of M-M M-M Theorem was initially meant for capital structure. But it applies to all aspects of financial policy: ¾ ¾ ¾ ¾ ¾ capital structure is irrelevant. long-term vs. short-term debt is irrelevant. dividend policy is irrelevant. risk management is irrelevant. etc. Indeed, the proof applies to all financial transactions because they are all zero NPV transactions. 23 Using M-M Sensibly M-M is not a literal statement about the real world. It obviously leaves important things out. But it gets you to ask the right question: How is this financing move going to change the size of the pie? M-M exposes some fallacies such as: ¾ WACC fallacy ¾ Win-Win fallacy ¾ EPS fallacy 24 WACC Fallacy: “Debt is Better Because Debt Is Cheaper Than Equity.” Because (for essentially all firms) debt is safer than equity, investors demand a lower return for holding debt than for holding equity. (True) The difference is significant: 4% vs. 13% expected return! So, companies should always finance themselves with debt because they have to give away less returns to investors, i.e., debt is cheaper. (False) What is wrong with this argument? 25 WACC Fallacy (cont.) This reasoning ignores the “hidden” cost of debt: ¾ Raising more debt makes existing equity more risky ¾ Is it still true when default probability is zero? Milk analogy: Whole milk = Cream + Skimmed milk People often confuse the two meanings of “cheap”: ¾ Low cost ¾ Good deal More on this in the “Valuation” module (Part II). 26 EPS Fallacy: “Debt is Better When It Makes EPS Go Up.” EPS can go up (or down) when a company increases its leverage. (True) Companies should choose their financial policy to maximize their EPS. (False) What is wrong with this argument? 27 EPS Fallacy (cont.) EBI(T) is unaffected by a change in capital structure (Recall that we assumed no taxes for now). Creditors receive the safe (or the safest) part of EBIT. Expected EPS might increase but EPS has become riskier! Remarks: Also tells us to be careful when using P/E ratios, e.g. comparing P/E ratios of companies with different capital structures. Further confusing effect in share-repurchases: The number of shares changes as well as expected earnings. 28 Leverage, returns, and risk Firm is a portfolio of debt and equity Assets Liab & Eq Long-Term Debt Net Assets Equity Therefore … D E r + rE rA = D A A and D E β + βE βA = D A A 29 Leverage, returns, and risk Asset risk is determined by the type of projects, not how the projects are financed Changes in leverage do not affect rA or βA Leverage affects rE and βE βA = D E βD + βE V V D β E = β A + (β A − β D ) E rA = D E rD + rE V V D rE = rA + (rA − rD ) E 30 Leverage and beta 4 βE Beta 3 2 βA 1 βD -1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Debt to equity ratio 31 Leverage and required returns 0.30 rE required return 0.25 0.20 rA 0.15 rD 0.10 0.05 0 0.2 0.4 0.6 0.8 Debt to equity ratio 1 1.2 1.4 32 Example Your firm is all equity financed and has $1 million of assets and 10,000 shares of stock (stock price = $100). Earnings before interest and taxes next year will be either $50,000, $125,000, or $200,000 depending on economic conditions. These earnings are expected to continue indefinitely. The payout ratio is 100%. The firm is thinking about a leverage recapitalization, selling $300,000 of debt and using the proceeds to repurchase stock. The interest rate is 10%. How would this transaction affect the firm’s EPS and stock price? Ignore taxes. 33 Current: all equity # of shares Debt Bad 10,000 $0 Expected 10,000 $0 Good 10,000 $0 EBIT Interest Net income EPS $50,000 0 $50,000 $5 $125,000 0 $125,000 $12.50 $200,000 0 $200,000 $20 Expected EPS = $12.5 Stock price = $100 rE = DPS / price = EPS / price = 12.5% 34 Recap: 30% debt # of shares Debt (r=10%) EBIT Interest Net income EPS Bad 7,000 $300,000 Expected 7,000 $300,000 Good 7,000 $300,000 $50,000 30,000 $20,000 $2.86 $125,000 30,000 $95,000 $13.57 $200,000 30,000 $170,000 $24.29 Expected EPS = $13.57 rE = rA + D/E (rA – rD) = 0.125 + (0.30/0.70) (0.125 – 0.10) = 13.57% Stock price = DPS / rE = EPS / rE = $100 35 Win-Win Fallacy: “Debt Is Better Because Some Investors Prefer Debt to Equity.” Investors differ in their preferences and needs, and thus want different cash flow streams. (True) Example: Young professionals vs. Retirees The sum of what all investors will pay is greater if the firm issues different securities (e.g., debt and equity) tailored for different clienteles of investors (Financial Marketing). (False) What is wrong with this argument? 36 Win-Win Fallacy (cont.) This reasoning assumes incomplete markets, i.e., that: ¾ There are indeed clienteles for different securities ¾ These clienteles are “unsatisfied”, i.e., that investors cannot replicate the security at the same or even lower cost. A large unsatisfied clientele for corporate debt is unlikely, as there exist close substitutes to any particular firm’s debt. Also, financial intermediaries are in the business of identifying unsatisfied clientele. Win-Win situation is more likely for more exotic securities or sophisticated financial arrangement 37 Practical Implications When evaluating a decision (e.g., the effect of a merger): → Separate financial (RHS) and real (LHS) parts of the move → MM tells that most value is created on LHS When evaluating an argument in favor of a financial decision: → Understand that it is wrong under MM assumptions → What departures from MM assumptions does it rely upon? → If none, then this is very dubious argument. → If some, try to assess their magnitude. 38 What’s Missing from the Simple M-M Story? Taxes: → Corporate taxes → Personal taxes Costs of Financial Distress 39 Capital Structure and Corporate Taxes Different financial transactions are taxed differently: → Interest payments are tax exempt for the firm. → Dividends and retained earnings are not. → Etc. Financial policy matters because it affects a firm’s tax bill. 40 Debt Tax Shield Claim: Debt increases firm value by reducing the tax burden. Example: XYZ Inc. generates a safe $100M annual perpetuity. Assume risk-free rate of 10%. Compare: ¾ 100% debt: perpetual $100M interest ¾ 100% equity: perpetual $100M dividend or capital gains Income before tax Corporate tax rate 35% Income after tax Firm value 100% Debt Interest Income $100M 0 $100M 100% Equity Equity income $100M -$35M $65M $1,000M $650M 41 Intuition MM still holds: The pie is unaffected by capital structure. Size of the pie = Value of before-tax cashflows But the IRS gets a slice too Financial policy affects the size of that slice. Interest payments being tax deductible, the PV of the IRS’ slice can be reduced by using debt rather than equity. 42 “Pie” Theory Equity Debt Taxes 43 Example In 2000, Microsoft had sales of $23 billion, earnings before taxes of $14.3 billion, and net income of $9.4 billion. Microsoft paid $4.9 billion in taxes, had a market value of $423 billion, and had no long-term debt outstanding. Bill Gates is thinking about a recapitalization, issuing $50 billion in long-term debt (rd = 7%) and repurchasing $50 billion in stock. How would this transaction affect Microsoft’s after-tax cashflows and shareholder wealth? 44 Microsoft: Balance sheet in $ millions Item Cash Current assets Current liabs LT debt Bk equity Mkt equity Sales EBIT Taxes Net income Oper CF 1997 8,966 10,373 3,610 0 9,797 155,617 1998 13,927 15,889 5,730 0 15,647 267,700 1999 17,236 20,233 8,718 0 27,485 460,770 2000 23,798 30,308 9,755 0 41,368 422,640 11,358 5,314 1,860 3,454 4,689 14,484 7,117 2,627 4,490 6,880 19,747 11,891 4,106 7,785 10,003 22,956 14,275 4,854 9,421 13,961 45 Microsoft, 2000 ($ millions) EBIT Interest (r × 50,000) Earnings before taxes Taxes (34%) After-tax earnings Cashflow to debtholders Cashflow to equityholders Total cashflows to D & E No Debt $14,275 0 $14,275 4,854 $9,421 Debt $14,275 3,500 $10,775 3,664 $7,111 $0 $9,421 $9,421 $3,500 $7,111 $10,611 46 Tax savings of debt Marginal tax rate = τ Taxes for unlevered firm………………τ EBIT Taxes for levered firm………………... .τ (EBIT – interest) Interest tax shield …………………..τ interest Interest = rd D Interest tax shield (each year) = τ rd D Note: only interest, not principal, payments reduce taxes 47